Signatures and labels in a blockchain derived from digital images
US-2024193394-A1 · Jun 13, 2024 · US
US9530171B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9530171-B2 |
| Application number | US-201114126604-A |
| Country | US |
| Kind code | B2 |
| Filing date | Aug 24, 2011 |
| Priority date | Aug 24, 2011 |
| Publication date | Dec 27, 2016 |
| Grant date | Dec 27, 2016 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A forensic authentication system includes an imaging device to capture an image of a printed mark and a non-printed area of a substrate directly adjacent to the printed mark, and a processor to run computer readable instructions. The processor can run computer readable instructions to utilize a model to define a substrate region that corresponds with at least a portion of the non-printed area of the substrate directly adjacent to the printed mark; and computer readable instructions to generate a substrate signature for the defined substrate region. Each of the computer readable instructions is embedded on a non-transitory, tangible computer readable medium.
Opening claim text (preview).
What is claimed is: 1. A forensic authentication system, comprising: an imaging device to capture an image of a printed mark and a non-printed area of a substrate outside of and directly adjacent to the printed mark; and a processor to run: computer readable instructions to determine a first plurality of points in the printed mark and a second plurality of corresponding points in the non-printed area based on predetermined information identifying the printed mark; computer readable instructions to, for each point in the first plurality, select a portion of the substrate between that point and a corresponding point in the second plurality; computer readable instructions to determine a region of the substrate that includes the portion of the substrate for each point in the first plurality; and computer readable instructions to generate a substrate signature for the region of the substrate based on a characteristic of the substrate inherent to the substrate, wherein the substrate signature is generated by calculating a measure of the characteristic at each of a plurality of predetermined locations in the region; wherein each of the computer readable instructions is embedded on a non-transitory, tangible computer readable medium. 2. The forensic authentication system as defined in claim 1 wherein the computer readable instructions to determine the region of the substrate include computer readable instructions to locate the printed mark in the image and align a previously stored outline of the printed mark with the printed mark; and wherein the predetermined information identifies the first and second plurality of points relative to a location of the outline. 3. The forensic authentication system as defined in claim 2 wherein the outline comprises coordinates of a plurality of contour points, the contour points indicative of an outer edge of the printed mark. 4. The forensic authentication system as defined in claim 3 wherein the computer readable instructions to determine the region of the substrate include: computer readable instructions to generate normal vectors for each contour point, the normal vectors perpendicular to the outer edge of the printed mark, wherein each point in the second plurality of points lies on a corresponding normal vector; computer readable instructions, for each point in the first plurality, to select a set of pixels between that point and the corresponding point in the second plurality as the portion of the substrate; computer readable instructions to construct a profile image by including each set of pixels as a column of pixels in the profile image, wherein the set of pixels are concatenated to form the column of pixels, and wherein the columns of pixels are concatenated to form the profile image; and computer readable instructions to select a predetermined portion of the profile image as the region of the substrate. 5. The forensic authentication system as defined in claim 4 wherein a height of the profile image is selected to be proportional to a size of the printed mark, and wherein the predetermined portion is a section of the profile image that is furthest from the printed mark and free of ink. 6. The forensic authentication system as defined in claim 1 wherein the processor further runs computer readable instructions to divide the substrate signature into a plurality of intervals, compute a statistical variance of the signature across each interval, and compute a coding of the substrate signature as a concatenation of the variances. 7. The forensic authentication system as defined in claim 1 wherein the processor further runs computer readable instructions to compute a distance function for comparing the substrate signature and another substrate signature. 8. The forensic authentication system as defined in claim 4 wherein the computer readable instructions to generate the substrate signature include computer readable instructions to compute an edge strength function to compare grayscale values for the pixels in each column, for each column, compute a sum of outputs from the edge strength function, and form the substrate signature from the sum for each column. 9. The forensic authentication system of claim 8 , wherein the computer readable instructions to compute an edge strength function for the pixels include computer readable instructions to compute a digital derivate of grayscale values of the pixels. 10. The forensic authentication system of claim 4 , wherein the normal vectors include at least two normal vectors non-parallel with each other, and wherein the profile image comprises a rectangular array of grayscale values. 11. A method to perform forensic authentication, comprising: receiving an image of a printed mark and a non-printed area of a substrate outside of and directly adjacent to the printed mark, wherein the printed mark comprises a character; determining a first plurality of points in the character and a second plurality of corresponding points in the non-printed area based on predetermined information identifying the character; for each point in the first plurality, selecting a portion of the substrate between that point and a corresponding point in the second plurality; selecting a region of the substrate that includes the portion of the substrate for each point in the first plurality; and generating a substrate signature for the region of the substrate based on a characteristic of the substrate inherent to the substrate, wherein the substrate signature is generated by calculating a measure of the characteristic at each of a plurality of predetermined locations in the region; wherein each element of the method is performed by a processor running computer readable instructions embedded on a non-transitory, tangible computer readable medium. 12. The method as defined in claim 11 wherein: determining the first and second plurality of points includes: finding a homographic transformation to align a plurality of contour points with an outer edge of the character, generating normal vectors for each contour point, the normal vectors perpendicular to the outer edge of the printed mark, and selecting each point in the first plurality of points as a point on a corresponding normal vector and each point in the second plurality of points as a point on the corresponding normal vector; selecting the portion comprises constructing a profile image by selecting the portion of the substrate for each point in the first plurality as a column of pixels in the profile image; and selecting the region comprises selecting a predetermined portion of the profile image free of ink as the region of the substrate. 13. The method as defined in claim 12 wherein the method further comprises low pass filtering the image. 14. The method as defined in claim 12 wherein generating the substrate signature includes computing an edge strength function for the pixels in each column, for each column, computing a sum of outputs from the edge strength function, and concatenating the sums for the columns to produce the substrate signature. 15. The method of claim 11 , wherein the character comprises an alphanumeric character. 16. The method of claim 12 , wherein selecting the predetermined portion of the profile image comprises selecting a predetermined percentage of the pixels in each column, wherein the selected pixels are farthest from the printed mark. 17. A non-transitory computer readable medium comprising instruction code, which when executed by a processor, causes the processor to: determine a first plurality of points in a printe
Checking or certification of the authentication information, e.g. by comparison with data stored independently · CPC title
of machine readable codes or marks, e.g. bar codes or glyphs · CPC title
Image watermarking · CPC title
Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title · CPC title
for tracing or tracking, e.g. forensic tracing of unauthorized copies · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.